| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.12, | |
| "eval_steps": 20, | |
| "global_step": 1140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.9520000000000002e-05, | |
| "loss": 0.675, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 2.904e-05, | |
| "loss": 0.6234, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.856e-05, | |
| "loss": 0.4842, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 2.8080000000000002e-05, | |
| "loss": 0.3626, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.7600000000000003e-05, | |
| "loss": 0.3259, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.712e-05, | |
| "loss": 0.2786, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6135040521621704, | |
| "eval_loss": 1.1749801635742188, | |
| "eval_runtime": 4843.6207, | |
| "eval_samples_per_second": 0.887, | |
| "eval_steps_per_second": 0.028, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 2.6640000000000002e-05, | |
| "loss": 0.231, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.616e-05, | |
| "loss": 0.2978, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 2.568e-05, | |
| "loss": 0.2383, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.52e-05, | |
| "loss": 0.2364, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 2.472e-05, | |
| "loss": 0.2186, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 2.4240000000000002e-05, | |
| "loss": 0.1869, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6817229390144348, | |
| "eval_loss": 0.9132505655288696, | |
| "eval_runtime": 4886.8402, | |
| "eval_samples_per_second": 0.879, | |
| "eval_steps_per_second": 0.028, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.3760000000000003e-05, | |
| "loss": 0.1656, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.328e-05, | |
| "loss": 0.1846, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.2800000000000002e-05, | |
| "loss": 0.1609, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2.232e-05, | |
| "loss": 0.1497, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 2.184e-05, | |
| "loss": 0.1564, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 2.136e-05, | |
| "loss": 0.1409, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6696158051490784, | |
| "eval_loss": 0.9002014398574829, | |
| "eval_runtime": 4890.7807, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.028, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 2.088e-05, | |
| "loss": 0.1441, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 2.04e-05, | |
| "loss": 0.1179, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 1.9920000000000002e-05, | |
| "loss": 0.1467, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 1.944e-05, | |
| "loss": 0.1311, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 1.896e-05, | |
| "loss": 0.1261, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.848e-05, | |
| "loss": 0.1533, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.1296, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7143189907073975, | |
| "eval_loss": 0.8478733897209167, | |
| "eval_runtime": 4805.5106, | |
| "eval_samples_per_second": 0.894, | |
| "eval_steps_per_second": 0.028, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 1.7519999999999998e-05, | |
| "loss": 0.1276, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 1.704e-05, | |
| "loss": 0.1259, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 1.656e-05, | |
| "loss": 0.1175, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 1.6080000000000002e-05, | |
| "loss": 0.1392, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 1.56e-05, | |
| "loss": 0.113, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 1.5120000000000001e-05, | |
| "loss": 0.1139, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7427241206169128, | |
| "eval_loss": 0.8348045349121094, | |
| "eval_runtime": 4819.2179, | |
| "eval_samples_per_second": 0.891, | |
| "eval_steps_per_second": 0.028, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 1.464e-05, | |
| "loss": 0.1246, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "learning_rate": 1.416e-05, | |
| "loss": 0.1313, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 1.3680000000000001e-05, | |
| "loss": 0.1175, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 1.32e-05, | |
| "loss": 0.1205, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 1.272e-05, | |
| "loss": 0.1015, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 1.224e-05, | |
| "loss": 0.0941, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7427241206169128, | |
| "eval_loss": 0.830384373664856, | |
| "eval_runtime": 4804.5403, | |
| "eval_samples_per_second": 0.894, | |
| "eval_steps_per_second": 0.028, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 1.1760000000000001e-05, | |
| "loss": 0.1224, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 1.128e-05, | |
| "loss": 0.105, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 1.08e-05, | |
| "loss": 0.122, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 1.032e-05, | |
| "loss": 0.1044, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 9.84e-06, | |
| "loss": 0.1077, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 9.36e-06, | |
| "loss": 0.1075, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6938300132751465, | |
| "eval_loss": 0.8423438668251038, | |
| "eval_runtime": 4824.348, | |
| "eval_samples_per_second": 0.89, | |
| "eval_steps_per_second": 0.028, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 8.88e-06, | |
| "loss": 0.0951, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 0.1229, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.1171, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 7.44e-06, | |
| "loss": 0.0955, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.1139, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 6.48e-06, | |
| "loss": 0.1051, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6e-06, | |
| "loss": 0.1011, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7236321568489075, | |
| "eval_loss": 0.8057857155799866, | |
| "eval_runtime": 4915.4234, | |
| "eval_samples_per_second": 0.874, | |
| "eval_steps_per_second": 0.027, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 5.52e-06, | |
| "loss": 0.0976, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 5.04e-06, | |
| "loss": 0.1008, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 4.56e-06, | |
| "loss": 0.1066, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 4.080000000000001e-06, | |
| "loss": 0.1038, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 3.6e-06, | |
| "loss": 0.1044, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 3.1199999999999998e-06, | |
| "loss": 0.1015, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7201396822929382, | |
| "eval_loss": 0.7934114336967468, | |
| "eval_runtime": 4842.9757, | |
| "eval_samples_per_second": 0.887, | |
| "eval_steps_per_second": 0.028, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 2.6399999999999997e-06, | |
| "loss": 0.1091, | |
| "step": 1140 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 1250, | |
| "num_train_epochs": 10, | |
| "save_steps": 20, | |
| "total_flos": 4.43240458915968e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |